Documento de resultados
Daniel Miranda, Juan Carlos Castillo, Catalina Miranda & José Conejeros
Abril, 2020
Load Data
Sample Size
#ICCS LA: 2009 - 2016
iccs_count <- mergeiccs %>%
group_by(time, ncountry) %>%
dplyr::summarise(N=n(), man=mean(s_gender, na.rm=T), age=mean(s_age, na.rm=T)) %>%
arrange(ncountry)
#ID School
iccs_count2 <- mergeiccs %>%
group_by(time, ncountry, idschool) %>%
dplyr::summarise(n=n()) %>%
dplyr::summarise(s=n()) %>%
arrange(ncountry)
iccs_count$s <- iccs_count2$s
rm(iccs_count2)
iccs_count$time <- as.character(iccs_count$time)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
kable(iccs_count, align = c("lcccccc")) %>%
kable_styling(bootstrap_options = c("striped", "hover"))| time | ncountry | N | man | age | s |
|---|---|---|---|---|---|
| 2009 | Chile | 5173 | 0.5143801 | 14.17872 | 177 |
| 2016 | Chile | 5081 | 0.4928164 | 14.17268 | 178 |
| 2009 | Colombia | 6200 | 0.5352932 | 14.37904 | 196 |
| 2016 | Colombia | 5609 | 0.5229096 | 14.59167 | 150 |
| 2009 | Dominican Republic | 4569 | 0.5468819 | 14.85543 | 145 |
| 2016 | Dominican Republic | 3937 | 0.5128270 | 14.18576 | 141 |
| 2009 | Guatemala | 3998 | 0.4899699 | 15.51745 | 145 |
| 2009 | Mexico | 6565 | 0.5222493 | 14.08043 | 215 |
| 2016 | Mexico | 5526 | 0.5000000 | 14.03109 | 213 |
| 2009 | Paraguay | 3391 | 0.5213801 | 14.81850 | 149 |
| 2016 | Peru | 5166 | 0.4816105 | 14.03021 | 206 |
Institutional Trust
#Design
mergesvy <- mergeiccs %>%
as_survey_design(
strata = jkzones,
weights = totwgts,
ids = jkreps,
nest = TRUE)
#Adjust and table
mergesvy[["variables"]][["nac_gob_d"]] <- as.character(mergesvy[["variables"]][["nac_gob_d"]])
mergesvy[["variables"]][["police_d"]] <- as.character(mergesvy[["variables"]][["police_d"]])
mergesvy[["variables"]][["pol_parties_d"]] <- as.character(mergesvy[["variables"]][["pol_parties_d"]])
mergesvy[["variables"]][["people_d"]] <- as.character(mergesvy[["variables"]][["people_d"]])
#########################
#National Goberment
#########################
table_freq_01 <- mergesvy %>%
dplyr::group_by(country, time, nac_gob_d) %>%
summarize(proportion = survey_mean(,na.rm=TRUE, "ci"))
#Table
#print(xtable(table_freq_01, caption = "Sample", format="text"), include.rownames=FALSE)
kable(table_freq_01, align = c("lcccccc")) %>%
kable_styling(bootstrap_options = c("striped", "hover"))| country | time | nac_gob_d | proportion | proportion_low | proportion_upp |
|---|---|---|---|---|---|
| CHL | 2009 | 0 | 0.3486123 | 0.3291055 | 0.3681191 |
| CHL | 2009 | 1 | 0.6513877 | 0.6318809 | 0.6708945 |
| CHL | 2009 | NA | NA | NA | NA |
| CHL | 2016 | 0 | 0.5041627 | 0.4847861 | 0.5235392 |
| CHL | 2016 | 1 | 0.4958373 | 0.4764608 | 0.5152139 |
| CHL | 2016 | NA | NA | NA | NA |
| COL | 2009 | 0 | 0.3797472 | 0.3552204 | 0.4042741 |
| COL | 2009 | 1 | 0.6202528 | 0.5957259 | 0.6447796 |
| COL | 2009 | NA | NA | NA | NA |
| COL | 2016 | 0 | 0.4476026 | 0.4236968 | 0.4715084 |
| COL | 2016 | 1 | 0.5523974 | 0.5284916 | 0.5763032 |
| COL | 2016 | NA | NA | NA | NA |
| DOM | 2009 | 0 | 0.2612638 | 0.2365092 | 0.2860184 |
| DOM | 2009 | 1 | 0.7387362 | 0.7139816 | 0.7634908 |
| DOM | 2009 | NA | NA | NA | NA |
| DOM | 2016 | 0 | 0.2219649 | 0.1992088 | 0.2447210 |
| DOM | 2016 | 1 | 0.7780351 | 0.7552790 | 0.8007912 |
| DOM | 2016 | NA | NA | NA | NA |
| GTM | 2009 | 0 | 0.5469156 | 0.5200227 | 0.5738086 |
| GTM | 2009 | 1 | 0.4530844 | 0.4261914 | 0.4799773 |
| GTM | 2009 | NA | NA | NA | NA |
| MEX | 2009 | 0 | 0.4158023 | 0.3954032 | 0.4362014 |
| MEX | 2009 | 1 | 0.5841977 | 0.5637986 | 0.6045968 |
| MEX | 2009 | NA | NA | NA | NA |
| MEX | 2016 | 0 | 0.4299568 | 0.4090000 | 0.4509136 |
| MEX | 2016 | 1 | 0.5700432 | 0.5490864 | 0.5910000 |
| MEX | 2016 | NA | NA | NA | NA |
| PER | 2016 | 0 | 0.5096989 | 0.4893269 | 0.5300710 |
| PER | 2016 | 1 | 0.4903011 | 0.4699290 | 0.5106731 |
| PER | 2016 | NA | NA | NA | NA |
| PRY | 2009 | 0 | 0.3409936 | 0.3156542 | 0.3663329 |
| PRY | 2009 | 1 | 0.6590064 | 0.6336671 | 0.6843458 |
| PRY | 2009 | NA | NA | NA | NA |
#########################
#Police
#########################
table_freq_02 <- mergesvy %>%
dplyr::group_by(country, time, police_d) %>%
summarize(proportion = survey_mean(,na.rm=TRUE, "ci"))
#Table
#print(xtable(table_freq_02, caption = "Sample", format="text"), include.rownames=FALSE)
kable(table_freq_02, align = c("lcccccc")) %>%
kable_styling(bootstrap_options = c("striped", "hover"))| country | time | police_d | proportion | proportion_low | proportion_upp |
|---|---|---|---|---|---|
| CHL | 2009 | 0 | 0.2896503 | 0.2717131 | 0.3075874 |
| CHL | 2009 | 1 | 0.7103497 | 0.6924126 | 0.7282869 |
| CHL | 2009 | NA | NA | NA | NA |
| CHL | 2016 | 0 | 0.3551635 | 0.3366103 | 0.3737167 |
| CHL | 2016 | 1 | 0.6448365 | 0.6262833 | 0.6633897 |
| CHL | 2016 | NA | NA | NA | NA |
| COL | 2009 | 0 | 0.4507952 | 0.4293250 | 0.4722654 |
| COL | 2009 | 1 | 0.5492048 | 0.5277346 | 0.5706750 |
| COL | 2009 | NA | NA | NA | NA |
| COL | 2016 | 0 | 0.5076088 | 0.4828832 | 0.5323344 |
| COL | 2016 | 1 | 0.4923912 | 0.4676656 | 0.5171168 |
| COL | 2016 | NA | NA | NA | NA |
| DOM | 2009 | 0 | 0.4369578 | 0.4107802 | 0.4631354 |
| DOM | 2009 | 1 | 0.5630422 | 0.5368646 | 0.5892198 |
| DOM | 2009 | NA | NA | NA | NA |
| DOM | 2016 | 0 | 0.4393891 | 0.4152856 | 0.4634927 |
| DOM | 2016 | 1 | 0.5606109 | 0.5365073 | 0.5847144 |
| DOM | 2016 | NA | NA | NA | NA |
| GTM | 2009 | 0 | 0.6664363 | 0.6423384 | 0.6905341 |
| GTM | 2009 | 1 | 0.3335637 | 0.3094659 | 0.3576616 |
| GTM | 2009 | NA | NA | NA | NA |
| MEX | 2009 | 0 | 0.5699461 | 0.5528957 | 0.5869966 |
| MEX | 2009 | 1 | 0.4300539 | 0.4130034 | 0.4471043 |
| MEX | 2009 | NA | NA | NA | NA |
| MEX | 2016 | 0 | 0.5097115 | 0.4933779 | 0.5260450 |
| MEX | 2016 | 1 | 0.4902885 | 0.4739550 | 0.5066221 |
| MEX | 2016 | NA | NA | NA | NA |
| PER | 2016 | 0 | 0.4978979 | 0.4810746 | 0.5147213 |
| PER | 2016 | 1 | 0.5021021 | 0.4852787 | 0.5189254 |
| PER | 2016 | NA | NA | NA | NA |
| PRY | 2009 | 0 | 0.5494438 | 0.5270717 | 0.5718159 |
| PRY | 2009 | 1 | 0.4505562 | 0.4281841 | 0.4729283 |
| PRY | 2009 | NA | NA | NA | NA |
#########################
#Political Parties
#########################
table_freq_03 <- mergesvy %>%
dplyr::group_by(country, time, pol_parties_d) %>%
summarize(proportion = survey_mean(,na.rm=TRUE, "ci"))
#Table
#print(xtable(table_freq_03, caption = "Sample", format="text"), include.rownames=FALSE)
kable(table_freq_03, align = c("lcccccc")) %>%
kable_styling(bootstrap_options = c("striped", "hover"))| country | time | pol_parties_d | proportion | proportion_low | proportion_upp |
|---|---|---|---|---|---|
| CHL | 2009 | 0 | 0.6550144 | 0.6346769 | 0.6753518 |
| CHL | 2009 | 1 | 0.3449856 | 0.3246482 | 0.3653231 |
| CHL | 2009 | NA | NA | NA | NA |
| CHL | 2016 | 0 | 0.6745075 | 0.6587782 | 0.6902368 |
| CHL | 2016 | 1 | 0.3254925 | 0.3097632 | 0.3412218 |
| CHL | 2016 | NA | NA | NA | NA |
| COL | 2009 | 0 | 0.6506877 | 0.6295206 | 0.6718549 |
| COL | 2009 | 1 | 0.3493123 | 0.3281451 | 0.3704794 |
| COL | 2009 | NA | NA | NA | NA |
| COL | 2016 | 0 | 0.7228073 | 0.7029239 | 0.7426906 |
| COL | 2016 | 1 | 0.2771927 | 0.2573094 | 0.2970761 |
| COL | 2016 | NA | NA | NA | NA |
| DOM | 2009 | 0 | 0.4887925 | 0.4644688 | 0.5131161 |
| DOM | 2009 | 1 | 0.5112075 | 0.4868839 | 0.5355312 |
| DOM | 2009 | NA | NA | NA | NA |
| DOM | 2016 | 0 | 0.5024324 | 0.4798019 | 0.5250629 |
| DOM | 2016 | 1 | 0.4975676 | 0.4749371 | 0.5201981 |
| DOM | 2016 | NA | NA | NA | NA |
| GTM | 2009 | 0 | 0.7368829 | 0.7169853 | 0.7567805 |
| GTM | 2009 | 1 | 0.2631171 | 0.2432195 | 0.2830147 |
| GTM | 2009 | NA | NA | NA | NA |
| MEX | 2009 | 0 | 0.6535926 | 0.6329986 | 0.6741866 |
| MEX | 2009 | 1 | 0.3464074 | 0.3258134 | 0.3670014 |
| MEX | 2009 | NA | NA | NA | NA |
| MEX | 2016 | 0 | 0.6261329 | 0.6059780 | 0.6462878 |
| MEX | 2016 | 1 | 0.3738671 | 0.3537122 | 0.3940220 |
| MEX | 2016 | NA | NA | NA | NA |
| PER | 2016 | 0 | 0.6660155 | 0.6471337 | 0.6848973 |
| PER | 2016 | 1 | 0.3339845 | 0.3151027 | 0.3528663 |
| PER | 2016 | NA | NA | NA | NA |
| PRY | 2009 | 0 | 0.6755932 | 0.6567009 | 0.6944855 |
| PRY | 2009 | 1 | 0.3244068 | 0.3055145 | 0.3432991 |
| PRY | 2009 | NA | NA | NA | NA |
#########################
#People in general
#########################
table_freq_04 <- mergesvy %>%
dplyr::group_by(country, time, people_d) %>%
summarize(proportion = survey_mean(,na.rm=TRUE, "ci"))
#Table
#print(xtable(table_freq_04, caption = "Sample", format="text"), include.rownames=FALSE)
kable(table_freq_04, align = c("lcccccc")) %>%
kable_styling(bootstrap_options = c("striped", "hover"))| country | time | people_d | proportion | proportion_low | proportion_upp |
|---|---|---|---|---|---|
| CHL | 2009 | 0 | 0.4825848 | 0.4652249 | 0.4999446 |
| CHL | 2009 | 1 | 0.5174152 | 0.5000554 | 0.5347751 |
| CHL | 2009 | NA | NA | NA | NA |
| CHL | 2016 | 0 | 0.5222508 | 0.5053029 | 0.5391987 |
| CHL | 2016 | 1 | 0.4777492 | 0.4608013 | 0.4946971 |
| CHL | 2016 | NA | NA | NA | NA |
| COL | 2009 | 0 | 0.5116344 | 0.4933059 | 0.5299628 |
| COL | 2009 | 1 | 0.4883656 | 0.4700372 | 0.5066941 |
| COL | 2009 | NA | NA | NA | NA |
| COL | 2016 | 0 | 0.5651447 | 0.5430788 | 0.5872106 |
| COL | 2016 | 1 | 0.4348553 | 0.4127894 | 0.4569212 |
| COL | 2016 | NA | NA | NA | NA |
| DOM | 2009 | 0 | 0.3897426 | 0.3636206 | 0.4158645 |
| DOM | 2009 | 1 | 0.6102574 | 0.5841355 | 0.6363794 |
| DOM | 2009 | NA | NA | NA | NA |
| DOM | 2016 | 0 | 0.3801493 | 0.3586919 | 0.4016067 |
| DOM | 2016 | 1 | 0.6198507 | 0.5983933 | 0.6413081 |
| DOM | 2016 | NA | NA | NA | NA |
| GTM | 2009 | 0 | 0.5258643 | 0.5041887 | 0.5475399 |
| GTM | 2009 | 1 | 0.4741357 | 0.4524601 | 0.4958113 |
| GTM | 2009 | NA | NA | NA | NA |
| MEX | 2009 | 0 | 0.5341219 | 0.5191580 | 0.5490859 |
| MEX | 2009 | 1 | 0.4658781 | 0.4509141 | 0.4808420 |
| MEX | 2009 | NA | NA | NA | NA |
| MEX | 2016 | 0 | 0.4829065 | 0.4638750 | 0.5019381 |
| MEX | 2016 | 1 | 0.5170935 | 0.4980619 | 0.5361250 |
| MEX | 2016 | NA | NA | NA | NA |
| PER | 2016 | 0 | 0.5284377 | 0.5113651 | 0.5455104 |
| PER | 2016 | 1 | 0.4715623 | 0.4544896 | 0.4886349 |
| PER | 2016 | NA | NA | NA | NA |
| PRY | 2009 | 0 | 0.4276067 | 0.4082165 | 0.4469968 |
| PRY | 2009 | 1 | 0.5723933 | 0.5530032 | 0.5917835 |
| PRY | 2009 | NA | NA | NA | NA |
Latin American 2009-2016
Support Authoritarianism
#Students’ endorsement of authoritarian government practices
#IRT WLE scores with mean of 10 and standard deviation of 2 within each participating country
dicta <- ggplot(mergeiccs, aes(civic_knowledge, l_autgov, weight=totwgts, linetype = country))
dicta <- dicta + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Support for authoritarian practices", colour = "dark")
dicta
#Dictatorships are justified when they bring order and safety
dicta1 <- ggplot(mergeiccs, aes(civic_knowledge, dicta_safety, weight=totwgts, linetype = country))
dicta1 <- dicta1 + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Support to dictatorship (order and safety)", colour = "dark")
dicta1
#Dictatorships are justified when they bring economic benefits
dicta2 <- ggplot(mergeiccs, aes(civic_knowledge, dicta_benefits, weight=totwgts, linetype = country))
dicta2 <- dicta2 + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Support to dictatorship (economic benefits)", colour = "dark")
dicta2
#All graphs
library(ggpubr)
png(file = "~/Dropbox/book_authoritarianism/Results/graph/g1.png", height = 350)
ggarrange(dicta, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2png(file = "~/Dropbox/book_authoritarianism/Results/graph/g2.png", height = 480)
ggarrange(dicta1, dicta2, ncol = 1, nrow = 2, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2Institutuional Trust
#Students' trust in civic institutions
#IRT WLE scores with mean of 10 and standard deviation of 2 within each participating country
#National Goberment
#Local Goberment
#Courts of justice
#The police
#National Parliament
#Trust in civic institutions
trust <- ggplot(mergeiccs, aes(civic_knowledge, s_intrust, weight=totwgts, linetype = country))
trust <- trust + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Students' trust in civic institutions", colour = "dark")
trust
#Trust in civic institutions
trust2 <- ggplot(mergeiccs, aes(l_autgov, s_intrust, weight=totwgts, linetype = country))
trust2 <- trust2 + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Authoritarian support", y = "Students' trust in civic institutions", colour = "dark")
trust2
#Trust in National Goberment
nac <- ggplot(mergeiccs, aes(civic_knowledge, nac_gob, weight=totwgts, linetype = country))
nac <- nac + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in National Goberment ", colour = "dark")
nac
#Trust in Local Goberment
local <- ggplot(mergeiccs, aes(civic_knowledge, local_gob, weight=totwgts, linetype = country))
local <- local + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in Local Goberment", colour = "dark")
local
#Trust National Parliament
parliament <- ggplot(mergeiccs, aes(civic_knowledge, parliament, weight=totwgts, linetype = country))
parliament <- parliament + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in National Parliament", colour = "dark")
parliament
#Political Parties
parties <- ggplot(mergeiccs, aes(civic_knowledge, pol_parties, weight=totwgts, linetype = country))
parties <- parties + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in Political Parties", colour = "dark")
parties
#Trust Courts of justice
courts <- ggplot(mergeiccs, aes(civic_knowledge, courts, weight=totwgts, linetype = country))
courts <- courts + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in Courts of justice", colour = "dark")
courts
#Trust The police
police <- ggplot(mergeiccs, aes(civic_knowledge, police, weight=totwgts, linetype = country))
police <- police + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in the Police", colour = "dark")
police
#Trust people in general
people <- ggplot(mergeiccs, aes(civic_knowledge, people, weight=totwgts, linetype = country))
people <- people + geom_smooth(method = "lm", alpha = 0.1, color="black", size=0.5) +
facet_wrap(~ time) +
labs(x = "Civic Knowledge", y = "Trust in People", colour = "dark")
people
#All graphs
png(file = "~/Dropbox/book_authoritarianism/Results/graph/g3.png", height = 350)
ggarrange(trust, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2png(file = "~/Dropbox/book_authoritarianism/Results/graph/g3b.png", height = 350)
ggarrange(trust2, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2png(file = "~/Dropbox/book_authoritarianism/Results/graph/g4.png", height = 480)
ggarrange(nac, local, ncol = 1, nrow = 2, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2png(file = "~/Dropbox/book_authoritarianism/Results/graph/g5.png", height = 480)
ggarrange(parliament, parties, ncol = 1, nrow = 2, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2png(file = "~/Dropbox/book_authoritarianism/Results/graph/g6.png", height = 480)
ggarrange(courts, police, ncol = 1, nrow = 2, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2png(file = "~/Dropbox/book_authoritarianism/Results/graph/g7.png", height = 350)
ggarrange(people, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()## quartz_off_screen
## 2Models and Interaction
#Modelos de regresión
#VD: Trust
#VD: Support authoritarianism
#VI: Trust
#VC: Nivel educacional de los padres, Libros en la casa, género del estudiante, nivel de discusión política, cohortes
##########################################################3
#OLS
##########################################################3
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ civic_knowledge , data=mergeiccs, w=totwgts)
m2 <- lm(l_autgov ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc, data=mergeiccs, w=totwgts)
m3 <- lm(l_autgov ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + factor(time) + factor(idcountry), data=mergeiccs, w=totwgts)
summary(m3)Call: lm(formula = l_autgov ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + factor(time) + factor(idcountry), data = mergeiccs, weights = totwgts)
Weighted Residuals: Min 1Q Median 3Q Max -1200.28 -35.35 6.71 44.62 825.45
Coefficients: Estimate Std. Error t value Pr(>|t|)
(Intercept) 85.4372913 0.3357651 254.456 < 0.0000000000000002 civic_knowledge -0.0793543 0.0005124 -154.854 < 0.0000000000000002 s_hisced 0.0457209 0.0285430 1.602 0.109200
s_homelit -0.2327802 0.0366456 -6.352 0.000000000214 s_gender -1.3150772 0.0768900 -17.103 < 0.0000000000000002 s_poldisc -0.0005511 0.0040444 -0.136 0.891606
factor(time)2016 1.0401339 0.0848272 12.262 < 0.0000000000000002 factor(idcountry)170 0.7941203 0.1685824 4.711 0.000002476532 factor(idcountry)214 -0.4167443 0.2477346 -1.682 0.092532 .
factor(idcountry)320 0.1032563 0.3191145 0.324 0.746264
factor(idcountry)484 0.5492984 0.1528144 3.595 0.000325 *** factor(idcountry)600 -0.4963251 0.3862910 -1.285 0.198851
factor(idcountry)604 0.0631990 0.2084167 0.303 0.761713
— Signif. codes: 0 ‘’ 0.001 ’’ 0.01 ’’ 0.05 ‘.’ 0.1 ’ ’ 1
Residual standard error: 99.04 on 52135 degrees of freedom (3067 observations deleted due to missingness) Multiple R-squared: 0.364, Adjusted R-squared: 0.3638 F-statistic: 2486 on 12 and 52135 DF, p-value: < 0.00000000000000022
(Intercept) 42.30073
(Intercept) 61.78515
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=mergeiccs, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + l_autgov , data=mergeiccs, w=totwgts)
m6 <- lm(s_intrust ~ civic_knowledge*l_autgov, data=mergeiccs, w=totwgts)
m7 <- lm(s_intrust ~ civic_knowledge*l_autgov + s_hisced + s_homelit + s_gender + s_poldisc, data=mergeiccs, w=totwgts)
m8 <- lm(s_intrust ~ civic_knowledge*l_autgov + s_hisced + s_homelit + s_gender + s_poldisc + factor(time) + factor(idcountry), data=mergeiccs, w=totwgts)
summary(m5)Call: lm(formula = s_intrust ~ civic_knowledge + l_autgov, data = mergeiccs, weights = totwgts)
Weighted Residuals: Min 1Q Median 3Q Max -1031.62 -54.23 -5.40 48.67 1078.49
Coefficients: Estimate Std. Error t value Pr(>|t|)
(Intercept) 42.8869120 0.5058082 84.79 <0.0000000000000002 civic_knowledge -0.0109151 0.0006818 -16.01 <0.0000000000000002 l_autgov 0.2332272 0.0050399 46.28 <0.0000000000000002 *** — Signif. codes: 0 ‘’ 0.001 ’’ 0.01 ’’ 0.05 ‘.’ 0.1 ’ ’ 1
Residual standard error: 114.5 on 51791 degrees of freedom (3421 observations deleted due to missingness) Multiple R-squared: 0.09054, Adjusted R-squared: 0.0905 F-statistic: 2578 on 2 and 51791 DF, p-value: < 0.00000000000000022
(Intercept) 39.4923
(Intercept) 36.7526
#Table
#library(texreg)
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3), digits = 2)
#texreg(list(m4, m5, m6, m7, m8), digits = 2)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 85.49*** | 85.93*** | 85.44*** | |
| (0.21) | (0.30) | (0.34) | ||
| civic_knowledge | -0.08*** | -0.08*** | -0.08*** | |
| (0.00) | (0.00) | (0.00) | ||
| s_hisced | -0.06* | 0.05 | ||
| (0.03) | (0.03) | |||
| s_homelit | -0.28*** | -0.23*** | ||
| (0.04) | (0.04) | |||
| s_gender | -1.37*** | -1.32*** | ||
| (0.08) | (0.08) | |||
| s_poldisc | 0.00 | -0.00 | ||
| (0.00) | (0.00) | |||
| factor(time)2016 | 1.04*** | |||
| (0.08) | ||||
| factor(idcountry)170 | 0.79*** | |||
| (0.17) | ||||
| factor(idcountry)214 | -0.42 | |||
| (0.25) | ||||
| factor(idcountry)320 | 0.10 | |||
| (0.32) | ||||
| factor(idcountry)484 | 0.55*** | |||
| (0.15) | ||||
| factor(idcountry)600 | -0.50 | |||
| (0.39) | ||||
| factor(idcountry)604 | 0.06 | |||
| (0.21) | ||||
| R2 | 0.36 | 0.36 | 0.36 | |
| Adj. R2 | 0.36 | 0.36 | 0.36 | |
| Num. obs. | 54769 | 52148 | 52148 | |
| RMSE | 99.05 | 99.25 | 99.04 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | ||
|---|---|---|---|---|---|---|
| (Intercept) | 63.10*** | 42.89*** | 13.74*** | 11.94*** | 13.20*** | |
| (0.26) | (0.51) | (1.25) | (1.28) | (1.29) | ||
| civic_knowledge | -0.03*** | -0.01*** | 0.05*** | 0.04*** | 0.04*** | |
| (0.00) | (0.00) | (0.00) | (0.00) | (0.00) | ||
| l_autgov | 0.23*** | 0.81*** | 0.75*** | 0.72*** | ||
| (0.01) | (0.02) | (0.02) | (0.02) | |||
| civic_knowledge:l_autgov | -0.00*** | -0.00*** | -0.00*** | |||
| (0.00) | (0.00) | (0.00) | ||||
| s_hisced | 0.02 | -0.02 | ||||
| (0.03) | (0.03) | |||||
| s_homelit | -0.31*** | -0.27*** | ||||
| (0.04) | (0.04) | |||||
| s_gender | -0.67*** | -0.73*** | ||||
| (0.09) | (0.09) | |||||
| s_poldisc | 0.11*** | 0.11*** | ||||
| (0.00) | (0.00) | |||||
| factor(time)2016 | 0.06 | |||||
| (0.10) | ||||||
| factor(idcountry)170 | -0.32 | |||||
| (0.19) | ||||||
| factor(idcountry)214 | 2.61*** | |||||
| (0.29) | ||||||
| factor(idcountry)320 | -3.47*** | |||||
| (0.37) | ||||||
| factor(idcountry)484 | -0.19 | |||||
| (0.18) | ||||||
| factor(idcountry)600 | -0.25 | |||||
| (0.46) | ||||||
| factor(idcountry)604 | -2.51*** | |||||
| (0.24) | ||||||
| R2 | 0.05 | 0.09 | 0.10 | 0.11 | 0.12 | |
| Adj. R2 | 0.05 | 0.09 | 0.10 | 0.11 | 0.12 | |
| Num. obs. | 51954 | 51794 | 51794 | 50281 | 50281 | |
| RMSE | 117.01 | 114.45 | 113.75 | 113.41 | 112.94 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||||
#Plot
library(sjPlot)
plot2 <- plot_model(m8, type = "pred", terms = c("civic_knowledge[300, 400, 500, 600, 700]", "l_autgov[30, 40, 50, 60 ,70]"), colors = 'bw', axis.title = c("Civic Knowledge", "Students' trust in civic institutions"), title = c(""), legend.title = "Support authoritarianism")
plot2
png(file = "~/Dropbox/book_authoritarianism/Results/graph/g8.png", height = 350)
ggarrange(plot2, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()quartz_off_screen 2
#Plot
plot3 <- plot_model(m8, type = "pred", terms = c("l_autgov[30, 40, 50, 60 ,70]", "civic_knowledge[300, 400, 500, 600, 700]"), colors = 'bw', axis.title = c("Support for authoritarian practices", "Students' trust in civic institutions"), title = c(""), legend.title = "Civic knowledge")
plot3
png(file = "~/Dropbox/book_authoritarianism/Results/graph/g8b.png", height = 350)
ggarrange(plot3, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()quartz_off_screen 2
OLS by chile
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
chi <- mergeiccs %>% filter(idcountry == 152)
#VD: Support authoritarianism
library(miceadds)## Loading required package: mice##
## Attaching package: 'mice'## The following object is masked from 'package:tidyr':
##
## complete## The following objects are masked from 'package:base':
##
## cbind, rbind## * miceadds 3.5-14 (2019-08-23 13:23:33)m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=chi, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=chi, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=chi, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=chi, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=chi, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 72.44*** | 793.05*** | 792.12*** | |
| (0.71) | (56.41) | (56.36) | ||
| s_intrust | 0.20*** | 0.18*** | 0.37*** | |
| (0.01) | (0.01) | (0.05) | ||
| civic_knowledge | -0.07*** | -0.07*** | -0.05*** | |
| (0.00) | (0.00) | (0.00) | ||
| s_hisced | -0.02 | -0.02 | ||
| (0.09) | (0.09) | |||
| s_homelit | -0.14 | -0.14 | ||
| (0.09) | (0.09) | |||
| s_gender | -0.67*** | -0.67*** | ||
| (0.18) | (0.18) | |||
| s_poldisc | 0.01 | 0.01 | ||
| (0.01) | (0.01) | |||
| time | -0.36*** | -0.36*** | ||
| (0.03) | (0.03) | |||
| s_intrust:civic_knowledge | -0.00*** | |||
| (0.00) | ||||
| R2 | 0.36 | 0.38 | 0.38 | |
| Adj. R2 | 0.36 | 0.38 | 0.38 | |
| Num. obs. | 10071 | 9855 | 9855 | |
| RMSE | 62.81 | 62.23 | 62.18 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 56.49*** | 949.93*** | |
| (0.61) | (64.44) | ||
| civic_knowledge | -0.02*** | -0.02*** | |
| (0.00) | (0.00) | ||
| s_hisced | -0.08 | ||
| (0.11) | |||
| s_homelit | -0.21* | ||
| (0.10) | |||
| s_gender | -0.86*** | ||
| (0.21) | |||
| s_poldisc | 0.13*** | ||
| (0.01) | |||
| time | -0.45*** | ||
| (0.03) | |||
| R2 | 0.02 | 0.05 | |
| Adj. R2 | 0.02 | 0.05 | |
| Num. obs. | 10089 | 9873 | |
| RMSE | 73.29 | 71.94 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
plot2 <- plot_model(m3, type = "pred", terms = c("s_intrust[20,80]", "civic_knowledge[200, 300, 400, 500, 600, 700, 800]"), colors = 'bw', axis.title = c("Students' trust in civic institutions", "Support for authoritarian practices"), title = c(""), legend.title = "Civic Knowledge")
plot2
OLS by Colombia
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
col <- mergeiccs %>% filter(idcountry == 170)
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=col, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=col, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=col, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=col, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=col, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 77.16*** | -268.35*** | -268.65*** | |
| (0.66) | (44.89) | (44.89) | ||
| s_intrust | 0.10*** | 0.11*** | 0.16*** | |
| (0.01) | (0.01) | (0.05) | ||
| civic_knowledge | -0.07*** | -0.07*** | -0.07*** | |
| (0.00) | (0.00) | (0.00) | ||
| s_hisced | 0.07 | 0.07 | ||
| (0.05) | (0.05) | |||
| s_homelit | -0.13 | -0.13 | ||
| (0.07) | (0.07) | |||
| s_gender | -1.28*** | -1.28*** | ||
| (0.15) | (0.15) | |||
| s_poldisc | -0.03*** | -0.03*** | ||
| (0.01) | (0.01) | |||
| time | 0.17*** | 0.17*** | ||
| (0.02) | (0.02) | |||
| s_intrust:civic_knowledge | -0.00 | |||
| (0.00) | ||||
| R2 | 0.35 | 0.36 | 0.36 | |
| Adj. R2 | 0.35 | 0.36 | 0.36 | |
| Num. obs. | 11233 | 11038 | 11038 | |
| RMSE | 83.19 | 82.73 | 82.72 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 62.34*** | 258.87*** | |
| (0.59) | (55.83) | ||
| civic_knowledge | -0.03*** | -0.03*** | |
| (0.00) | (0.00) | ||
| s_hisced | 0.24*** | ||
| (0.06) | |||
| s_homelit | -0.24** | ||
| (0.09) | |||
| s_gender | -1.80*** | ||
| (0.18) | |||
| s_poldisc | 0.16*** | ||
| (0.01) | |||
| time | -0.10*** | ||
| (0.03) | |||
| R2 | 0.04 | 0.08 | |
| Adj. R2 | 0.04 | 0.08 | |
| Num. obs. | 11262 | 11064 | |
| RMSE | 105.09 | 103.12 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
plot2 <- plot_model(m3, type = "pred", terms = c("s_intrust[20,80]", "civic_knowledge[200, 300, 400, 500, 600, 700, 800]", "time"), colors = 'bw', axis.title = c("Students' trust in civic institutions", "Support for authoritarian practices"), title = c(""), legend.title = "Civic Knowledge")
plot2
OLS by Dominican Republic
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
dom <- mergeiccs %>% filter(idcountry == 214)
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=dom, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=dom, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=dom, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=dom, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=dom, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 74.97*** | -169.07** | -169.51** | |
| (0.85) | (60.43) | (60.43) | ||
| s_intrust | 0.13*** | 0.12*** | 0.17** | |
| (0.01) | (0.01) | (0.05) | ||
| civic_knowledge | -0.07*** | -0.07*** | -0.07*** | |
| (0.00) | (0.00) | (0.01) | ||
| s_hisced | 0.03 | 0.03 | ||
| (0.08) | (0.08) | |||
| s_homelit | -0.44*** | -0.44*** | ||
| (0.10) | (0.10) | |||
| s_gender | -0.85*** | -0.85*** | ||
| (0.21) | (0.21) | |||
| s_poldisc | -0.02 | -0.02 | ||
| (0.01) | (0.01) | |||
| time | 0.12*** | 0.12*** | ||
| (0.03) | (0.03) | |||
| s_intrust:civic_knowledge | -0.00 | |||
| (0.00) | ||||
| R2 | 0.31 | 0.32 | 0.32 | |
| Adj. R2 | 0.31 | 0.32 | 0.32 | |
| Num. obs. | 7117 | 6655 | 6655 | |
| RMSE | 48.31 | 47.95 | 47.95 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 71.67*** | -228.12** | |
| (0.75) | (79.93) | ||
| civic_knowledge | -0.04*** | -0.04*** | |
| (0.00) | (0.00) | ||
| s_hisced | -0.12 | ||
| (0.10) | |||
| s_homelit | -0.17 | ||
| (0.13) | |||
| s_gender | -1.58*** | ||
| (0.27) | |||
| s_poldisc | 0.09*** | ||
| (0.01) | |||
| time | 0.15*** | ||
| (0.04) | |||
| R2 | 0.07 | 0.09 | |
| Adj. R2 | 0.07 | 0.09 | |
| Num. obs. | 7160 | 6685 | |
| RMSE | 64.17 | 63.62 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
plot2 <- plot_model(m3, type = "pred", terms = c("s_intrust[20,80]", "civic_knowledge[200, 300, 400, 500, 600, 700, 800]", "time"), colors = 'bw', axis.title = c("Students' trust in civic institutions", "Support for authoritarian practices"), title = c(""), legend.title = "Civic Knowledge")
plot2
OLS by Mexico
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
mex <- mergeiccs %>% filter(idcountry == 484)
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=mex, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=mex, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=mex, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=mex, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=mex, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 77.58*** | -267.31*** | -276.62*** | |
| (0.71) | (50.95) | (50.97) | ||
| s_intrust | 0.20*** | 0.19*** | 0.39*** | |
| (0.01) | (0.01) | (0.05) | ||
| civic_knowledge | -0.08*** | -0.08*** | -0.06*** | |
| (0.00) | (0.00) | (0.01) | ||
| s_hisced | 0.03 | 0.03 | ||
| (0.06) | (0.06) | |||
| s_homelit | -0.21** | -0.22** | ||
| (0.08) | (0.08) | |||
| s_gender | -1.05*** | -1.05*** | ||
| (0.17) | (0.17) | |||
| s_poldisc | -0.02* | -0.02* | ||
| (0.01) | (0.01) | |||
| time | 0.17*** | 0.17*** | ||
| (0.03) | (0.03) | |||
| s_intrust:civic_knowledge | -0.00*** | |||
| (0.00) | ||||
| R2 | 0.40 | 0.41 | 0.41 | |
| Adj. R2 | 0.40 | 0.41 | 0.41 | |
| Num. obs. | 11583 | 11379 | 11379 | |
| RMSE | 166.24 | 165.07 | 164.96 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 63.18*** | -272.13*** | |
| (0.57) | (58.97) | ||
| civic_knowledge | -0.03*** | -0.03*** | |
| (0.00) | (0.00) | ||
| s_hisced | -0.04 | ||
| (0.07) | |||
| s_homelit | -0.34*** | ||
| (0.09) | |||
| s_gender | -0.74*** | ||
| (0.20) | |||
| s_poldisc | 0.10*** | ||
| (0.01) | |||
| time | 0.16*** | ||
| (0.03) | |||
| R2 | 0.05 | 0.06 | |
| Adj. R2 | 0.05 | 0.06 | |
| Num. obs. | 11634 | 11426 | |
| RMSE | 193.85 | 191.71 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
plot2 <- plot_model(m3, type = "pred", terms = c("s_intrust[20,80]", "civic_knowledge[200, 300, 400, 500, 600, 700, 800]", "time"), colors = 'bw', axis.title = c("Students' trust in civic institutions", "Support for authoritarian practices"), title = c(""), legend.title = "Civic Knowledge")
plot2
OLS by Guatemala 2009
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
gtm <- mergeiccs %>% filter(idcountry == 320)
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=gtm, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=gtm, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=gtm, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=gtm, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=gtm, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 74.49*** | 75.83*** | 89.85*** | |
| (1.11) | (1.32) | (3.62) | ||
| s_intrust | 0.11*** | 0.11*** | -0.19* | |
| (0.01) | (0.01) | (0.07) | ||
| civic_knowledge | -0.07*** | -0.07*** | -0.10*** | |
| (0.00) | (0.00) | (0.01) | ||
| s_hisced | 0.23** | 0.26*** | ||
| (0.08) | (0.08) | |||
| s_homelit | -0.13 | -0.13 | ||
| (0.11) | (0.11) | |||
| s_gender | -1.34*** | -1.36*** | ||
| (0.24) | (0.24) | |||
| s_poldisc | -0.00 | 0.00 | ||
| (0.01) | (0.01) | |||
| s_intrust:civic_knowledge | 0.00*** | |||
| (0.00) | ||||
| R2 | 0.35 | 0.36 | 0.36 | |
| Adj. R2 | 0.35 | 0.36 | 0.36 | |
| Num. obs. | 3798 | 3684 | 3684 | |
| RMSE | 41.28 | 41.24 | 41.15 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 65.52*** | 58.11*** | |
| (0.93) | (1.35) | ||
| civic_knowledge | -0.04*** | -0.04*** | |
| (0.00) | (0.00) | ||
| s_hisced | -0.17 | ||
| (0.10) | |||
| s_homelit | -0.26 | ||
| (0.14) | |||
| s_gender | -0.74* | ||
| (0.30) | |||
| s_poldisc | 0.12*** | ||
| (0.02) | |||
| R2 | 0.10 | 0.12 | |
| Adj. R2 | 0.10 | 0.12 | |
| Num. obs. | 3804 | 3690 | |
| RMSE | 52.42 | 51.66 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading
OLS by Paraguay 2009
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
pry <- mergeiccs %>% filter(idcountry == 604)
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=pry, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=pry, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=pry, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=pry, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=pry, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 69.95*** | 71.15*** | 71.25*** | |
| (0.84) | (1.02) | (2.71) | ||
| s_intrust | 0.10*** | 0.09*** | 0.09 | |
| (0.01) | (0.01) | (0.05) | ||
| civic_knowledge | -0.05*** | -0.06*** | -0.06*** | |
| (0.00) | (0.00) | (0.01) | ||
| s_hisced | 0.35*** | 0.35*** | ||
| (0.08) | (0.08) | |||
| s_homelit | -0.04 | -0.04 | ||
| (0.10) | (0.10) | |||
| s_gender | -1.67*** | -1.67*** | ||
| (0.20) | (0.20) | |||
| s_poldisc | -0.00 | -0.00 | ||
| (0.01) | (0.01) | |||
| s_intrust:civic_knowledge | 0.00 | |||
| (0.00) | ||||
| R2 | 0.35 | 0.36 | 0.36 | |
| Adj. R2 | 0.35 | 0.36 | 0.36 | |
| Num. obs. | 5034 | 4914 | 4914 | |
| RMSE | 69.71 | 69.44 | 69.45 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 63.12*** | 55.90*** | |
| (0.68) | (1.08) | ||
| civic_knowledge | -0.03*** | -0.03*** | |
| (0.00) | (0.00) | ||
| s_hisced | 0.15 | ||
| (0.11) | |||
| s_homelit | 0.19 | ||
| (0.14) | |||
| s_gender | -1.75*** | ||
| (0.27) | |||
| s_poldisc | 0.14*** | ||
| (0.01) | |||
| R2 | 0.09 | 0.12 | |
| Adj. R2 | 0.09 | 0.11 | |
| Num. obs. | 5037 | 4916 | |
| RMSE | 92.75 | 91.50 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading
OLS by Peru 2016
##########################################################3
#OLS
##########################################################3
#Borramos los modelos generados en el apartado anterior
rm(list=(ls()[!ls() %in% ("mergeiccs")]))
per <- mergeiccs %>% filter(idcountry == 600)
#VD: Support authoritarianism
m1 <- lm(l_autgov ~ s_intrust + civic_knowledge, data=per, w=totwgts)
m2 <- lm(l_autgov ~ s_intrust + civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=per, w=totwgts)
m3 <- lm(l_autgov ~ s_intrust*civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=per, w=totwgts)
#VD: Trust
m4 <- lm(s_intrust ~ civic_knowledge, data=per, w=totwgts)
m5 <- lm(s_intrust ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + time, data=per, w=totwgts)
#Table
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1, m2, m3, m4, m5), digits = 2)
#htmlreg(list(m1, m2, m3, m4, m5), digits = 4)
htmlreg(list(m1, m2, m3), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 70.30*** | 71.54*** | 76.89*** | |
| (1.19) | (1.40) | (4.16) | ||
| s_intrust | 0.13*** | 0.10*** | -0.01 | |
| (0.01) | (0.02) | (0.08) | ||
| civic_knowledge | -0.06*** | -0.06*** | -0.07*** | |
| (0.00) | (0.00) | (0.01) | ||
| s_hisced | 0.15 | 0.15 | ||
| (0.10) | (0.10) | |||
| s_homelit | -0.02 | -0.01 | ||
| (0.14) | (0.14) | |||
| s_gender | -1.40*** | -1.41*** | ||
| (0.28) | (0.28) | |||
| s_poldisc | 0.01 | 0.01 | ||
| (0.01) | (0.01) | |||
| s_intrust:civic_knowledge | 0.00 | |||
| (0.00) | ||||
| R2 | 0.36 | 0.36 | 0.36 | |
| Adj. R2 | 0.35 | 0.36 | 0.36 | |
| Num. obs. | 2958 | 2756 | 2756 | |
| RMSE | 36.43 | 36.54 | 36.54 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | ||
|---|---|---|---|
| (Intercept) | 63.22*** | 59.21*** | |
| (0.92) | (1.28) | ||
| civic_knowledge | -0.03*** | -0.03*** | |
| (0.00) | (0.00) | ||
| s_hisced | 0.05 | ||
| (0.12) | |||
| s_homelit | 0.23 | ||
| (0.17) | |||
| s_gender | -2.32*** | ||
| (0.34) | |||
| s_poldisc | 0.10*** | ||
| (0.02) | |||
| R2 | 0.07 | 0.10 | |
| Adj. R2 | 0.07 | 0.10 | |
| Num. obs. | 2968 | 2766 | |
| RMSE | 45.62 | 44.53 | |
| p < 0.001, p < 0.01, p < 0.05 | |||
## Warning in predict.lm(model, newdata = fitfram, type = "response", se.fit =
## se, : prediction from a rank-deficient fit may be misleading
Multilevel Models Aproximation: cluster by schools
#Modelos de regresión
#VD: Trust
#VD: Support authoritarianism
#VI: Trust
#VC: Nivel educacional de los padres, Libros en la casa, género del estudiante, nivel de discusión política, cohortes
##########################################################3
#OLS
##########################################################3
#VD: Support authoritarianism
m1b <- lmer(l_autgov ~ civic_knowledge + (1|idschool), data=mergeiccs, w=totwgts)## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, : Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, : Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, : Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?#m4b <- lmer(l_autgov ~ civic_knowledge + s_hisced + s_homelit + s_gender + s_poldisc + ( 1|time) + ( 1|country) + (1|idschool), data=mergeiccs, w=totwgts)
#VD: Trust
m4b <- lmer(s_intrust ~ civic_knowledge + (1|idschool), data=mergeiccs, w=totwgts)## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, : Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, : Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?## Warning: Some predictor variables are on very different scales: consider rescaling
## Warning: Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?## Warning: Some predictor variables are on very different scales: consider rescaling
## Warning: Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?## Warning: Some predictor variables are on very different scales: consider rescaling
## Warning: Model is nearly unidentifiable: very large eigenvalue
## - Rescale variables?#m9b <- lmer(s_intrust ~ civic_knowledge*l_autgov + s_hisced + s_homelit + s_gender + s_poldisc + + (1|time) + (1|idcountry) + (1|idschool), data=mergeiccs, w=totwgts)
#Table
#library(texreg)
#print(xtable(iccs_count[, c(2:1,6,3:5)], caption = "Sample", format="text"), include.rownames=FALSE)
#texreg(list(m1b, m2b, m3b), digits = 2)
#texreg(list(m4b, m5b, m6b, m7b, m8b, m9b), digits = 2)
htmlreg(list(m1b, m2b, m3b), caption = "VD: Support authoritarianism")| Model 1 | Model 2 | Model 3 | ||
|---|---|---|---|---|
| (Intercept) | 85.56*** | 86.01*** | 85.63*** | |
| (0.24) | (0.31) | (0.35) | ||
| civic_knowledge | -0.08*** | -0.08*** | -0.08*** | |
| (0.00) | (0.00) | (0.00) | ||
| s_hisced | -0.05 | 0.04 | ||
| (0.03) | (0.03) | |||
| s_homelit | -0.24*** | -0.20*** | ||
| (0.04) | (0.04) | |||
| s_gender | -1.38*** | -1.32*** | ||
| (0.08) | (0.08) | |||
| s_poldisc | 0.00 | -0.00 | ||
| (0.00) | (0.00) | |||
| factor(time)2016 | 0.92*** | |||
| (0.09) | ||||
| factor(idcountry)170 | 0.80*** | |||
| (0.17) | ||||
| factor(idcountry)214 | -0.58* | |||
| (0.25) | ||||
| factor(idcountry)320 | -0.09 | |||
| (0.32) | ||||
| factor(idcountry)484 | 0.50** | |||
| (0.15) | ||||
| factor(idcountry)600 | -0.73 | |||
| (0.38) | ||||
| factor(idcountry)604 | 0.14 | |||
| (0.21) | ||||
| AIC | 420567.91 | 399727.19 | 399554.50 | |
| BIC | 420603.55 | 399798.09 | 399687.42 | |
| Log Likelihood | -210279.95 | -199855.60 | -199762.25 | |
| Num. obs. | 54769 | 52148 | 52148 | |
| Num. groups: idschool | 243 | 243 | 243 | |
| Var: idschool (Intercept) | 2.16 | 2.09 | 2.02 | |
| Var: Residual | 9554.61 | 9597.23 | 9563.16 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||
| Model 1 | Model 2 | Model 3 | Model 4 | Model 5 | ||
|---|---|---|---|---|---|---|
| (Intercept) | 62.99*** | 43.37*** | 14.16*** | 12.45*** | 13.68*** | |
| (0.28) | (0.52) | (1.26) | (1.29) | (1.30) | ||
| civic_knowledge | -0.03*** | -0.01*** | 0.05*** | 0.04*** | 0.04*** | |
| (0.00) | (0.00) | (0.00) | (0.00) | (0.00) | ||
| l_autgov | 0.23*** | 0.81*** | 0.75*** | 0.72*** | ||
| (0.01) | (0.02) | (0.02) | (0.02) | |||
| civic_knowledge:l_autgov | -0.00*** | -0.00*** | -0.00*** | |||
| (0.00) | (0.00) | (0.00) | ||||
| s_hisced | 0.02 | -0.02 | ||||
| (0.03) | (0.03) | |||||
| s_homelit | -0.29*** | -0.25*** | ||||
| (0.04) | (0.04) | |||||
| s_gender | -0.68*** | -0.74*** | ||||
| (0.09) | (0.09) | |||||
| s_poldisc | 0.10*** | 0.11*** | ||||
| (0.00) | (0.00) | |||||
| factor(time)2016 | 0.02 | |||||
| (0.10) | ||||||
| factor(idcountry)170 | -0.31 | |||||
| (0.19) | ||||||
| factor(idcountry)214 | 2.67*** | |||||
| (0.29) | ||||||
| factor(idcountry)320 | -3.39*** | |||||
| (0.37) | ||||||
| factor(idcountry)484 | -0.25 | |||||
| (0.18) | ||||||
| factor(idcountry)600 | -0.39 | |||||
| (0.46) | ||||||
| factor(idcountry)604 | -2.47*** | |||||
| (0.24) | ||||||
| AIC | 415561.93 | 412156.29 | 411539.04 | 398745.91 | 398345.59 | |
| BIC | 415597.36 | 412200.57 | 411592.17 | 398834.17 | 398495.62 | |
| Log Likelihood | -207776.96 | -206073.15 | -205763.52 | -199362.96 | -199155.79 | |
| Num. obs. | 51954 | 51794 | 51794 | 50281 | 50281 | |
| Num. groups: idschool | 243 | 243 | 243 | 243 | 243 | |
| Var: idschool (Intercept) | 2.12 | 1.63 | 1.59 | 1.53 | 1.49 | |
| Var: Residual | 13434.24 | 12896.01 | 12739.18 | 12669.48 | 12565.38 | |
| p < 0.001, p < 0.01, p < 0.05 | ||||||
#Plot
library(sjPlot)
plot2 <- plot_model(m8b, type = "pred", terms = c("civic_knowledge[300, 400, 500, 600, 700]", "l_autgov[30, 40, 50, 60 ,70]"), colors = 'bw', axis.title = c("Civic Knowledge", "Students' trust in civic institutions"), title = c(""), legend.title = "Support authoritarianism")
plot2
png(file = "~/Dropbox/book_authoritarianism/Results/graph/g9.png", height = 350)
ggarrange(plot2, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()quartz_off_screen 2
#Plot
plot3 <- plot_model(m8b, type = "pred", terms = c("l_autgov[30, 40, 50, 60 ,70]", "civic_knowledge[300, 400, 500, 600, 700]"), colors = 'bw', axis.title = c("Support for authoritarian practices", "Students' trust in civic institutions"), title = c(""), legend.title = "Civic knowledge")
plot3
png(file = "~/Dropbox/book_authoritarianism/Results/graph/g9b.png", height = 350)
ggarrange(plot3, ncol = 1, nrow = 1, common.legend = TRUE, legend="right")
dev.off()quartz_off_screen 2